Natural Abundance 17O DNP Two-Dimensional and Surface

Communication pubs.acs.org/JACS

Natural Abundance 17O DNP Two-Dimensional and SurfaceEnhanced NMR Spectroscopy Frédéric A. Perras,† Takeshi Kobayashi,† and Marek Pruski*,†,‡ †

U.S. DOE Ames Laboratory, Ames, Iowa 50011-3020, United States Department of Chemistry, Iowa State University, Ames, Iowa 50011-3020, United States

‡

S Supporting Information *

dynamic nuclear polarization (DNP).3 In DNP, the nuclear magnetization is enhanced by transferring polarization from unpaired electrons to the nuclei of interest by the application of high-power microwaves near the electron Larmor frequency.4 This approach can yield signal-to-noise enhancements up to γe/ γn, i.e., 658 for 1H and 4855 for 17O, potentially surpassing the sensitivity that can even be obtained by 17O isotopic enrichment alone. A particularly exciting application of DNP is for the selective enhancement of surface species that are often responsible for most of the chemistry of materials. This has been termed DNP surface-enhanced NMR spectroscopy (DNP-SENS)5 and has been applied to characterizing surface-supported catalysts and nanoparticles. The extension of DNP-SENS to include 17O NMR is highly desirable, for example, due to the importance of oxide supports in heterogeneous catalysis. However, further advances in sensitivity are necessary to enable the detection of such weak signals. In the studies of small nanoparticles,3 or 17O-enriched molecular species in solutions, for which the 17O spins can be in proximity to the radicals,6 17O nuclei have been directly polarized by DNP. Direct 17O DNP has also been performed in single crystals containing intrinsic paramagnetic impurities.7 Most recently, direct 17O DNP enhancements of as high as 115 have been obtained in biological solids using trityl radicals.6 However, direct 17O DNP of nonenriched samples has a major drawback in that spin diffusion within a material is too slow for the entire sample to be polarized.8 In samples containing hydrogen, it is advantageous to hyperpolarize 1H, utilize 1H−1H spin diffusion, and subsequently transfer the 1H magnetization to 17O nuclei within the bulk.3,9 The signal-to-noise enhancements that can be obtained using modern polarizing agents are also much larger when polarizing 1H than 17O;10 note that an enhancement as low as 40 on 1H translates to an enhancement of 295 for 17O, if we assume perfect polarization transfer. The transfer of magnetization from 1H to a quadrupolar nucleus, such as 17O, by cross-polarization (CP) presents, however, considerable challenges due to the unfavorable spin dynamics during spin locking and the CP process itself.11 The poor spin locking characteristics of the central transition (m = 1/2 to −1/2 transition) of a quadrupolar nucleus under MAS conditions generate severe line shape distortions,12 hampering the analysis of the spectra, and often also leads to sensitivity losses as opposed to gains.13 In order to take full advantage of the

ABSTRACT: Due to its extremely low natural abundance and quadrupolar nature, the 17O nuclide is very rarely used for spectroscopic investigation of solids by NMR without isotope enrichment. Additionally, the applicability of dynamic nuclear polarization (DNP), which leads to sensitivity enhancements of 2 orders of magnitude, to 17O is wrought with challenges due to the lack of spin diffusion and low polarization transfer efficiency from 1H. Here, we demonstrate new DNP-based measurements that extend 17 O solid-state NMR beyond its current capabilities. The use of the PRESTO technique instead of conventional 1 H−17O cross-polarization greatly improves the sensitivity and enables the facile measurement of undistorted line shapes and two-dimensional 1H−17O HETCOR NMR spectra as well as accurate internuclear distance measurements at natural abundance. This was applied for distinguishing hydrogen-bonded and lone 17O sites on the surface of silica gel; the one-dimensional spectrum of which could not be used to extract such detail. Lastly, this greatly enhanced sensitivity has enabled, for the first time, the detection of surface hydroxyl sites on mesoporous silica at natural abundance, thereby extending the concept of DNP surface-enhanced NMR spectroscopy to the 17O nuclide.

T

he chemistry of oxygen is ubiquitous in all areas of the chemical sciences, from oxide-based materials, such as oxide catalysts and ceramics, to biochemistry, where hydrogen bonding to oxygen plays an important role in protein folding, for example. Nonetheless, while the solid-state nuclear magnetic resonance (SSNMR) spectroscopy of the other elements of importance in the chemical sciences, such as 1H, 6/7Li, 11B, 13C, 15N, 27Al, 29Si, and 31P, to name a few, has flourished into a prominent characterization technique, 17O SSNMR investigations have failed to gain widespread use. The lagging development in 17O SSNMR spectroscopy is mainly due to the low natural isotopic abundance of 17O nuclei (NA = 0.038%), which renders the NMR signals unobservable unless tremendous experiment time is invested.1 Additional challenge is posed by the quadrupolar nature of the 17O nuclide (I = 5/2), which results in line broadening that cannot be removed by magic angle spinning (MAS). Most 17O SSNMR studies have then opted for isotopic enrichment,2 which is not always possible. Recently, however, it was demonstrated that 17O SSNMR spectra could be acquired in a reasonable amount of time by using © XXXX American Chemical Society

Received: April 15, 2015

A

DOI: 10.1021/jacs.5b03905 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Communication

Journal of the American Chemical Society

to a depletion of the signals at lower frequency due to the poor spin locking of those isochromats, whereas PRESTO yields the undistorted line shape and has an enhanced sensitivity, by a factor of 3. We then acquired natural abundance 17O SSNMR spectra of Mg(OH)2 and Ca(OH)2 using both the CP-QCPMG and PRESTO-QCPMG techniques (Figure 2). The samples were impregnated with a 16 mM 1,1,2,2-tetrachloroethane solution of the state-of-the-art TEKPol biradical17 and spun to 12.5 kHz at a temperature of 105 K. Note that in these two samples, the 1H−1H spin diffusion propagated the magnetization into the bulk of the materials. As can be seen, the PRESTO technique provided signal enhancements on the order of 2−11, depending on the spikelet, and an overall sensitivity enhancement of 5, when compared to CP. This improvement means that DNP-enhanced 17O SSNMR experiments at natural abundance that would take a day to perform using CP can be performed in an hour using PRESTO. Additionally, both singularities of the powder pattern can be observed by PRESTO, something that cannot be done using the CP spectra, enabling the determination of the chemical shift and quadrupolar coupling parameters;18 this is more clearly seen in the spectra reconstructed from the weighed sum of the echoes.19 The intensities of the singularities can also be used to extract information regarding the relative orientation of the dipolar and EFG tensors.14b SIMPSON20 simulations of the PRESTO spectra using the known chemical shift and EFG tensor parameters21 are also shown in Figure 2 and are in good agreement with the experiment. The PRESTO-QCPMG pulse sequence can also be modified to perform 1H−17O heteronuclear correlation (HETCOR) experiments (Figure 1c),22 where frequency-switched Lee− Goldburg homonuclear decoupling is applied during t1 to enhance 1H resolution.23 We have indeed acquired the 1H−17O HETCOR spectra for Mg(OH)2 and Ca(OH)2, at natural abundance, using this approach. The spectra in Figure 3 show that the 17O signal is mostly correlated to the hydroxyl resonance (located at 0 and 1 ppm for Mg(OH)2 and Ca(OH)2, respectively)24 and that polarization is not transferred directly from the solvent (C2H2Cl4). This is to be expected since PRESTO is mainly sensitive to short internuclear distances.14 17O−1H correlation spectra have also been acquired using 1H detection in

Figure 1. Pulse sequences to acquire natural abundance 17O NMR spectra with DNP: (a) CP-QCPMG, (b) PRESTO-QCPMG, and (c) PRESTO-QCPMG-HETCOR.

sensitivity improvements afforded by DNP, it is then necessary to utilize polarization transfer methods that overcome these two main issues associated with CP. The phase-shifted recoupling effects a smooth transfer of order (PRESTO) polarization transfer technique is such a method (see Figure 1 for the pulse sequence diagram).14 The PRESTO pulse sequence discards the problematic 17O spin locking pulse, and instead symmetry-based singlequantum heteronuclear recoupling is applied to 1H spins, while a simple Hahn echo is performed at the 17O Larmor frequency. In our case, the R1817 heteronuclear recoupling sequence was used,15 which also performs homonuclear decoupling, and the sensitivity is further enhanced by the application of quadrupolar Carr−Purcell−Meiboom−Gill (QCPMG) detection.16 Note, however, that alternative recoupling sequences may be used at higher MAS frequencies, such as R1825, to reduce the RF power requirements (Table S1). Additional details about the pulse sequences and the setup of all experiments are given in Supporting Information. A comparison of the 17O CP-QCPMG and PRESTO-QCPMG SSNMR spectra of an 17O-enriched sample of silica gel are shown in Figure S1. It can be seen that, as is well documented,12 CP leads

Figure 2. Natural-abundance, DNP-enhanced, 17O MAS NMR spectra of Mg(OH)2 (a−c) and Ca(OH)2 (d−f) acquired using the CP-QCPMG (c and f) and the PRESTO-QCPMG schemes (b and e). Both the spikelet (bottom) and reconstructed (top) spectra are shown. Signal to noise improvements from 2 to 11 (depending on the spikelet) are observed using PRESTO. SIMPSON simulations of the PRESTO MAS spectra are shown in (a) and (d). All spectra were acquired in 3.4 h. B

DOI: 10.1021/jacs.5b03905 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Communication

Journal of the American Chemical Society

respectively, correspond to O−H bond lengths of 0.97 ± 0.02 and 0.98 ± 0.02 Å. These values are in remarkable agreement with the neutron diffraction structures of 0.96 and 0.94 Å for Mg(OH)2 and Ca(OH)2, respectively.25 A similar approach has been used to study hydrogen bonding to oxygen in organic species.26 In those cases elongated O−H bond lengths were measured due to the effects of vibrational averaging at room temperature. When this experiment was performed on a 17O-enriched silica gel, the oscillations could not be simulated using only a single dipolar coupling constant (see Figure 4c). Instead, the curve corresponds to a sum of signals arising from hydrogen-bonded and lone surface silanol groups, having effective dipolar coupling constants of 17 ± 1 and 8.5 ± 1 kHz. The 50% reduction in dipolar coupling strength for some sites likely originates from partial averaging of the dipolar coupling under librational motions;27 similar effects have been seen using deuterium NMR of surface silanol groups in silica gel at these temperatures.28 In fact, such a model would indeed yield the fitted curve in Figure 4c. This type of motion is not structurally permitted in Mg(OH)2 and Ca(OH)2. The signals arising from the strongly dipolar coupled pairs must then represent rigid silanols, whose librations are restricted by the frozen solvent or by weak hydrogen bonding to neighboring silanol groups or to water molecules at the surface. In principle, the reduction in dipolar coupling could also originate from a large elongation of the O−H bond to 1.2 Å from a very strong hydrogen bond; this is, however, unlikely due to the prevalence of the site and the rarity of such interactions. The ability to study 1H−17O dipolar oscillations is very significant because this technique affords a straightforward way of studying the interactions between the surface of a silica support and guest molecules, even for multiple chemical environments with similar chemical shift and EFG parameters.2d Due to the remarkable sensitivity afforded by DNP-enhanced PRESTO-QCPMG NMR, we were able to perform the first natural abundance 17O DNP-SENS experiment.5 The 17O DNPSENS NMR spectrum of SBA-15, a mesoporous silica sample having a surface area of 600 m2/g and a pore diameter of 9.6 Å,29 acquired in 14 h is shown in Figure 5. This spectrum demonstrates

Figure 3. Natural-abundance, DNP-enhanced, 1H−17O MAS PRESTOQCPMG-HETCOR NMR spectra of Mg(OH)2 (top) and Ca(OH)2 (bottom) acquired in 4.5 h. Projections along the 1H dimension show that polarization is transferred mainly from the hydroxyl proton and not the solvent.

enriched samples;2d however, this would lead to t1 noise at natural abundance, preventing the detection of 17O. By incrementing either of the two dipolar recoupling periods (τ0 or τ1, see Figure 1b) in the PRESTO experiment, while keeping the other fixed, it is also possible to easily measure the 1 H−17O dipolar coupling and thus determine the internuclear distance.14 By decreasing the QCPMG period, so as to collapse all the signal intensity into a single spikelet, and incrementing the first recoupling block (τ0), so as to maintain an echo delay of a single rotor period, PRESTO dipolar oscillations were measured in 30 min for Mg(OH)2 and Ca(OH)2 at natural abundance (Figure 4a,b). The resulting 1H−17O dipolar coupling constants (RDD) of 18 ± 1 and 17 ± 1 kHz for Mg(OH)2 and Ca(OH)2,

Figure 5. Natural-abundance DNP-SENS 17O MAS NMR spectrum of the mesoporous silica SBA-15, acquired using the PRESTO-QCPMG pulse sequence in 14 h.

that by utilizing state-of-the-art equipment, biradicals, and pulse sequences, it will be possible to measure 1H−17O internuclear distances of catalyst supports, thereby enabling the characterization of hydrogen bonding and dynamics on the surface of catalytically important materials at natural abundance. In summary, the scope of natural abundance DNP-enhanced 17 O SSNMR is greatly enhanced via the application of improved

Figure 4. DNP-enhanced 1H−17O PRESTO-QCPMG dipolar oscillations from Mg(OH)2, Ca(OH)2, and 17O-enriched silica gel. SIMPSON simulations of the oscillations using the dipolar coupling strength indicated on the figure are shown in red. In (c), a sum of curves corresponding to lone and hydrogen-bonded silanol groups was inferred. C

DOI: 10.1021/jacs.5b03905 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Communication

Journal of the American Chemical Society

Bodenhausen, G.; Copéret, C.; Emsley, L. J. Am. Chem. Soc. 2010, 132, 15459. (b) Lelli, M.; Gajan, D.; Lesage, A.; Caporini, M. A.; Vitzthum, V.; Miéville, P.; Héroguel, F.; Rascón, F.; Roussey, A.; Thieuleux, C.; Boualleg, M.; Veyre, L.; Bodenhausen, G.; Copéret, C.; Emsley, L. J. Am. Chem. Soc. 2011, 133, 2104. (c) Wolf, P.; Valla, M.; Rossini, A. J.; ComasVives, A.; Núñez-Zarur, F.; Malaman, B.; Lesage, A.; Emsley, L.; Copéret, C.; Hermans, I. Angew. Chem., Int. Ed. 2014, 53, 10179. (d) Protesescu, L.; Rossini, A. J.; Kriegner, D.; Valla, M.; de Kergommeaux, A.; Walter, M.; Kravchyk, K. V.; Nachtegaal, M.; Stangl, J.; Malaman, B.; Reiss, P.; Lesage, A.; Emsley, L.; Copéret, C.; Kovalenko, M. V. ACS Nano 2014, 8, 2639. (e) Rossini, A. J.; Zagdoun, A.; Lelli, M.; Lesage, A.; Copéret, C.; Emsley, L. Acc. Chem. Res. 2013, 46, 1942. (6) Michaelis, V. K.; Corzilius, B.; Smith, A. A.; Griffin, R. G. J. Phys. Chem. B 2013, 117, 14894. (7) Brun, E.; Derighetti, B.; Hundt, E. E.; Niebuhr, H. H. Phys. Lett. 1970, 31A, 416. (8) (a) Lafon, O.; Thankamony, A. S. L.; Kobayashi, T.; Carnevale, D.; Vitzthum, V.; Slowing, I. I.; Kandel, K.; Vezin, H.; Amoureux, J.-P.; Bodenhausen, G.; Pruski, M. J. Phys. Chem. C 2013, 117, 1375. (b) Rossini, A. J.; Zagdoun, A.; Hegner, F.; Schwarzwälder, M.; Gajan, D.; Copéret, C.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2012, 134, 16899. (9) Michaelis, V. K.; Markhasin, E.; Daviso, E.; Herzfeld, J.; Griffin, R. G. J. Phys. Chem. Lett. 2012, 3, 2030. (10) Kubicki, D. J.; Rossini, A. J.; Purea, A.; Zagdoun, A.; Ouari, O.; Tordo, P.; Engelke, F.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2014, 136, 15711. (11) Vega, A. J. Solid State Nucl. Magn. Reson. 1992, 1, 17. (12) Barrie, P. J. Chem. Phys. Lett. 1993, 208, 486. (13) Pruski, M.; Lang, D. P.; Fernandez, C.; Amoureux, J. P. Solid State Nucl. Magn. Reson. 1997, 7, 327. (14) (a) Zhao, X.; Hoffbauer, W.; auf der Günne, J. S.; Levitt, M. H. Solid State Nucl. Magn. Reson. 2004, 26, 57. (b) van Beek, J. D.; Dupree, R.; Levitt, M. H. J. Magn. Reson. 2006, 179, 38. (c) Haies, I. M.; Jarvis, J. A.; Bentley, H.; Heinmaa, I.; Kuprov, I.; Williamson, P. T. F.; Carravetta, M. Phys. Chem. Chem. Phys. 2015, 17, 6577. (15) Zhao, X.; Sudmeier, J. L.; Bachovchin, W. W.; Levitt, M. H. J. Am. Chem. Soc. 2001, 123, 11097. (16) Larsen, F. H.; Skibsted, J.; Jakobsen, H. J.; Nielsen, N. C. J. Am. Chem. Soc. 2000, 122, 7080. (17) Zagdoun, A.; Casano, G.; Ouari, O.; Schwarzwälder, M.; Rossini, A. J.; Aussenac, F.; Yulikov, M.; Jeschke, G.; Copéret, C.; Lesage, A.; Tordo, P.; Emsley, L. J. Am. Chem. Soc. 2013, 135, 12790. (18) Field, T. R.; Bain, A. D. Solid State Nucl. Magn. Reson. 2014, 63−64, 42. (19) Lefort, R.; Wiench, J. W.; Pruski, M.; Amoureux, J.-P. J. Chem. Phys. 2002, 116, 2493. (20) Bak, M.; Rasmussen, J. T.; Nielsen, N. C. J. Magn. Reson. 2000, 147, 296. (21) Nyfeler, D.; Armbruster, T. Am. Mineral. 1998, 83, 119. (22) Sardo, M.; Siegel, R.; Santos, S. M.; Rocha, J.; Gomes, J. R. B.; Mafra, L. J. Phys. Chem. A 2012, 116, 6711. (23) Bielecki, A.; Kolbert, A. C.; Levitt, M. H. Chem. Phys. Lett. 1989, 155, 341. (24) Xue, X.; Kanzaki, M. J. Phys. Chem. B 2007, 111, 13156. (25) (a) Desgranges, L.; Calvarin, G.; Chevrier, G. Acta. Cryst. B 1996, 52, 82. (b) Busing, W. R.; Levy, H. A. J. Chem. Phys. 1957, 26, 563. (26) (a) Brinkmann, A.; Kentgens, A. P. M. J. Am. Chem. Soc. 2006, 128, 14758. (b) Brinkmann, A.; Kentgens, A. P. M. J. Phys. Chem. B 2006, 110, 16089. (27) Schanda, P.; Huber, M.; Boisbouvierm, J.; Meier, B. H.; Ernst, M. Angew. Chem., Int. Ed. 2011, 50, 11005. (28) Kobayashi, T.; DiVerdi, J. A.; Maciel, G. E. J. Phys. Chem. C 2008, 112, 4315. (29) Kandel, K.; Anderegg, J. W.; Nelson, N. C.; Chaudhary, U.; Slowing, I. I. J. Catal. 2014, 314, 142.

polarization transfer. The improved sensitivity of PRESTO decreases experiment times by over an order of magnitude, when compared to CP, such that 17O DNP NMR is now sufficiently sensitive for the application of 17O DNP-SENS at natural isotopic abundance. Additionally, reliable line shapes can be obtained from the spectra, and the experiment can be trivially altered to measure HETCOR spectra and 1H−17O internuclear distances. This has enabled the detection of hydrogen-bonded and non-hydrogenbonded silanol groups, whose dynamic motions can be probed by this approach, at the surface of silica gel. Such distinction cannot be made otherwise. These developments may facilitate the widespread application of 17O DNP-SENS NMR, as well as other techniques relying on transfer of polarization between spin-1/2 and quadrupolar nuclei, for probing the structure and dynamics in a variety of materials.

■

ASSOCIATED CONTENT

S Supporting Information *

The PRESTO-QCPMG pulse sequence is described. Details regarding the experiment setup, the simulations, and the sample preparations are given. The spectra acquired for an 17O-enriched silica gel are shown. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/ jacs.5b03905.

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank M. Carravetta and I. M. Haies for sharing their PRESTO-II pulse program, I. I. Slowing for synthesizing silica SBA-15, and J. Stebbins for 17O-enriched silica gel sample. This research is supported by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences, Division of Materials Science and Engineering (methods development), and Division of Chemical Sciences, Geosciences, and Biosciences (specific applications). Support for F.P. is through a Spedding Fellowship funded by the LDRD program. Ames Laboratory is operated for the DOE by Iowa State University under Contract No. DE-AC0207CH11358.

■

REFERENCES

(1) Ashbrook, S. E.; Farnan, I. Solid State Nucl. Magn. Reson. 2004, 26, 105. (2) (a) Wu, G.; Dong, S.; Ida, R.; Reen, N. J. Am. Chem. Soc. 2002, 124, 1768. (b) Zhu, J.; Ye, E.; Terskikh, V.; Wu, G. Angew. Chem., Int. Ed. 2010, 49, 8399. (c) Leskes, M.; Drewett, N. E.; Hardwick, L. J.; Bruce, P. G.; Goward, G. R.; Grey, C. P. Angew. Chem., Int. Ed. 2012, 51, 8560. (d) Merle, N.; Trébosc, J.; Baudouin, A.; Del Rosal, I.; Maron, L.; Szeto, K.; Genelot, M.; Mortreux, A.; Taoufik, M.; Delevoye, L.; Gauvin, R. M. J. Am. Chem. Soc. 2012, 134, 9263. (e) He, P.; Xu, J.; Terskikh, V. V.; Sutrisno, A.; Nie, H.-Y.; Huang, Y. J. Phys. Chem. C 2013, 117, 16953. (f) Kong, X.; O’Dell, L. A.; Terskikh, V.; Ye, E.; Wang, R.; Wu, G. J. Am. Chem. Soc. 2012, 134, 14609. (3) Blanc, F.; Sperrin, L.; Jefferson, D. A.; Pawsey, S.; Rosay, M.; Grey, C. P. J. Am. Chem. Soc. 2013, 135, 2975. (4) Maly, T.; Debelouchina, G. T.; Bajaj, V. S.; Hu, K.-N.; Joo, C.-G.; Mak-Jurkauskas, M. L.; Sirigiri, J. R.; van der Wel, P. C. A.; Herzfeld, J.; Temkin, R. J.; Griffin, R. G. J. Chem. Phys. 2008, 128, 052211. (5) (a) Lesage, A.; Lelli, M.; Gajan, D.; Caporini, M. A.; Vitzthum, V.; Miéville, P.; Alauzun, J.; Roussey, A.; Thieuleux, C.; Mehdi, A.; D

DOI: 10.1021/jacs.5b03905 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX